Surely life at Caltech isn't that bad?

The ants are back again and it's only going to get worse over summer.

We have ant poison in the TCS lab. Feel free to use it.

 [Alastair, Zach]

Today we vented the chamber and obstructed the cavity internally to measure the open-loop noise on the PDs themselves. To account for noise in both PDs while still rejecting anything that was common-mode---things like laser noise that are suppressed by the primary loop and absent from the gyro signal---I took the demodulated outputs of the PDs (i.e. the open-loop error signals) and subtracted them with an SR560. I then relocked the cavity, measured the current gyro noise spectrum, and took OLTFs to obtain optical responses with which to calibrate the error signals into gyro noise. The noise is far too low to explain the excess LF gyro noise:

I also measured the error signal spectra individually. Both of them are at about the same level as the differential noise, so there isn't a huge common-mode component.

As far as I can tell, we still haven't found the source of the blasted LF junk.

[Alastair, Zach]

Over the past couple days (in between overhauling malfunctioning PDs), we have been trying to hunt down the excess low-frequency noise. Last week, with Koji's help, we essentially ruled out the possibility of scattered light noise from the transmitted end of the gyro by intentionally reflecting varying amounts of light back into the cavity and observing NO DIFFERENCE in the low-frequency spectrum.

AM from the EOM still seems to be the most likely suspect, and we continue to find ways in which it could couple. Yesterday, we think we traced the crazy AM level modulation that I mentioned in my last post to the iris we are using to isolate the proper beam out of the AOM setup. The double-1st order (desired) and single-1st order (undesired) beams are extremely difficult to isolate from one another, so we had to use a very small aperture that we think was coupling EOM jitter into AM quite strongly. We moved the iris down the beam path closer to the waist(s) and were able to get better isolation with a larger aperture. The drift in AM level now seems absent.

While the AM levels in the CCW and CW beams (as measured on the RF analyzer) are not simultaneously minimized at the same pre-EOM HWP angle, there is no longer a large-angle discrepancy; the difference is <1 degree and the noise in both beams can now be kept fairly low at some compromise angle. I have set up a PDA255 in each path immediately before the cavity to monitor them both simultaneously. By adjusting the HWP before the EOM and fine-tuning the EOM orientation, I was able to get the AM peaks in both beams to be <10 ppm relative to the carrier.

After re-locking the gyro, I saw NO IMPROVEMENT in the low-frequency noise yet again. I replaced the PDA255s and tried looking at the LF noise directly after demodulation. I saw some excess noise above the broadband floor below about 1 Hz, which is roughly where the excess gyro noise begins. Upon deliberately de-tuning the HWP so that the RF peaks were >20x higher, however, I saw NO NOTICEABLE CHANGE in the audio spectrum. Looking at the peaks in the RF spectrum more closely, I discovered that the linewidths are below 1 Hz, suggesting that this may not be the source after all.

I suggest that we vent the chamber and obstruct the cavity so that we can run the same tests with the REFL PDs themselves. Then, we can use the noise we measure in the error signal to easily and accurately estimate the contribution to the NB.

 I got around to doing the EOM AM/PM measurement, and the results aren't fantastic.

PM:

I already measured this the other day, but we've done some switching stuff around since then so I did it again. 

• Carrier: 780 mV
• SB: 11.4 mV
• Γ ~ 2*sqrt(11.4/780) = 0.24 rads (this agrees with the other day's measurement)

AM:

This is where it gets messy. I measured this in two places: after the BS in the CCW path, and after the AOM in the CW path. It was measured by putting the signal from a PDA255 directly into the RF analyzer and measuring the peak (in V) at 19 MHz. The DC level of the light for all measurements was 3.00 V.

CCW:

The first thing I noticed was that the peak was not minimized at the same HWP angle that made the beam have exactly S polarization. Rather, it was best about 5 degrees off.

Even at this minimal point, the peak fluctuated in amplitude from ~10 uV to 180 uV with a period of 1-2 minutes. You could literally sit there without touching it and watch it go up and down before your eyes. WTF?

Rotating the beam to exactly S made the peak climb to 500 uV!

CW: 

This was even worse than the other direction. I had to turn it 17 degrees away from S in order to minimize the peak at ~50 uV. Putting it at S made it ~1.3 mV!

It didn't seem to oscillate as wildly as the CCW beam at the minimal point.

I think the most alarming thing is that the AM in both beams cannot be simultaneously minimized. This explains why we always see junk at low frequencies, I think: minimizing the peak for the CCW beam just ensures that there's more junk in the CW direction.

Need to figure out what gives.

I tuned up our first TRANS PD, S/N 03, that Alastair was working on with the J-Laser tonight. Here is the TF:

As before for the REFL diodes, I measured the output node of the diode with a probe so that I could tune the readout notch to 100 MHz. I had to change C2 to 33 pF to get 100 MHz in range with the tunable cap (C4), which I changed to a 1.5-50 pF one. Here is a full-span plot and a zoom around the notch:

I think it looks ratty at high frequencies because we are starting to see interference effects from the O(1 m) length cables.

 To aid my memory, the new values for the 100MHz notch

• C2=82pF
• C4=11.8pF(1.5-15pF tunable)
• L3=27nH

 So here is the outline of the Trans PD component choice:

1) Shot noise:  at 3mW the shot noise limit is ~3e-11A.

2) The transimpedance is entirely set by the component choices we make for the inductor/capacitor pair in the readout notch, since the diode capacitance and resistance are fixed.

3) We want to make sure we're well away from railing the opamp.  We want approx 0.3V output max, which is a transimpedance of 100.  This includes the gain of 10 from the opamp stage.

4) We are well below any noise level that could be of concern for us in transmission readout.

Since we are well above the noise level we care about in transmission, I've chosen the values to make sure the opamp isn't railing at 3mW input power.  This does mean we will be dominated by voltage noise in the opamp.

Based on this I picked the value of L (which had to be a value we have, 27nH) to give the nearest transimpedance to 10V/A which I checked in Matlab (see attached Matlab graph).  I then checked the model in LISO, and get the correct transimpedance for the full circuit including the opamp.  As you can see the noise is now dominated by voltage noise in the LMH6624, simply because we have such low optical gain.  It seems like the LMH6624 is not the nicest way to do this if we ever want to use more power than 3mW....  the minimum gain of 10 is causing us to reduce the gain at the point where we really want it - the input.

The noise level from this at 100MHz is 1e-10A/rt(Hz), which corresponds to a frequency noise of 3e-9 Hz / rt(Hz) at 0.1Hz.  The equivalent rotation noise is then 1e-15 rads / rt(Hz) at 0.1Hz

 I mentioned in the last post that I had measured the AM in the CCW beam after the beam splitter using a PDA255. I saw that the noise spectrum was flat well above where we see the excess low-frequency noise, so I ruled it out as the cause. What I didn't do is the same thing for the CW direction. I had a hunch today that AM could be generated by jitter of the input beam into the AOM caused by the EOM (Rana said that jitter was one source of EOM->AM coupling).

I put a PDA255 in the CW path after the AOM double-pass, stuck the signal into the demod setup, and---lo and behold---I saw low-frequency noise with just about the same shape we're seeing in the gyro signal:

I decided to try putting separate EOMs in each path. This was extremely messy for a few reasons:

• sheer lack of space: not only did I need room for the EOMs, but also for several waveplates to rotate the beam to S and then back to P into the faradays.
• inability to put in focusing lenses to get the beam through the crystal without fudging the whole modematching setup.
• only having one multi-axis EOM mount---one direction had to use one of the old fixed brass mounts, and I had to use what little parameter space I had to work with before the beam goes through the faraday isolator to do it.

Anyway, I found a way to do it that sort-of-kinda-maybe fit:

I split the power from the FG to the EOMs and realigned the beams into the cavity. Needless to say, the setup was far from optimal. Anyway, I decided to lock the cavity and PLL and see if I saw any improvement. There was none; in fact, the noise was worse. I then realized that I had forgotten to put the top on the box. I replaced it, and the noise went down a bit, but still slightly higher than before. I THEN realized that I didn't have the auxiliary box around the CCW injection steering mirrors. I replaced it, and the noise level was at just about where it was before. See below:

It is obvious that the sensitivity is suffering au cause d'the shitty way I put it together. I don't think we're seeing oscillator noise in the middle band since the PLL signal looks no better than the AOM signal (see previous post). I think it may be worth trying to do a better job at setting up the twin-EOM scheme, but there are two things that have me on the fence:

1. Doing this right essentially means redesigning the optical layout altogether. There are too may waveplates to count at the moment, and I'm certain that with a few minutes' thought we can come up with a better layout. We will absolutely need another multi-axis EOM mount to avoid over-constraining the beam path, and we don't have one. I think Frank and Dmass have one, so perhaps we can snag theirs and order another one, but this is time and money. The bright side to this is that we will be replacing the cavity optics either way, and it might be "therapeutic" to rebuild the IO anyway.
2. This is the tricky one: why do we see an equal improvement in both signals with the replacement of the box cover? No one ordered that, so to speak; the model predicts that IO noise should be suppressed by the CW loop gain, so the PLL should always look better than the AOM in bands where we're limited by IO noise. It could be that the AOM loop has $&#@all gain with my piss-poor attempt, but I think it's unlikely that we've lost over 60 dB of gain, even in this state.

I'll see what kind of loop diagnostics I can do in this limp-mode, and otherwise we'll just have to take a leap of faith that things will look better once we've reconfigured. If not, it will be the same amount of work again just to bring it back to the single-EOM setup. In the meantime, I'll try to think more about how the IO noise could couple to the PLL unsuppressed.

http://nodus.ligo.caltech.edu:8080/TCS_Lab/135

It does seem to suggest that the source of the low frequency noise is something that is not unique to either the transmission or reflection readout (ie not the MZ phase noise or PLL phase noise), and is not something that should be suppressed by the loop gain (like input optics noise or AOM oscillator phase noise) in the transmission readout.

EDIT: I should have mentioned that rotating the polarization before the EOM in order to minimize the error signal offset did not have a noticeable effect on the bulk of the low frequency noise. It did, however, make the coherent-looking peaks from 100-600 mHz go away, though I have no materials to support this.

Also, by placing a PDA255 after the initial beam splitter and demodulating it at the sideband frequency, I observed that the RFAM noise is flat well above the frequencies where we see the excess noise. This isn't purest way to rule it out, but I think it is convincing: if the noise we see at low frequencies is from RFAM, we would see it extending to higher frequencies.

 Below are the (delayed) results of Friday's oscillator noise measurements, calibrated to gyro units and plotted alongside the gyro noise. The traces are:

1. Gyro noise from AOM actuation
2. Gyro noise from PLL actuation
3. Two marconis with f_carr = 100 MHz and dev = 100 kHz/V beating together, with the beatnote fed back to one and the other's dev input shorted. This is divided by sqrt(2) to estimate the noise from the PLL oscillator alone in gyro mode. It agrees pretty well with (2) in the region we think is dominated by oscillator noise.
4. Two marconis with f_carr = 50 MHz and dev = 100 kHz/V beating together, with the beatnote fed back to one and the other's dev input shorted. This is multiplied by sqrt(2) (i.e., 2/sqrt(2)) to estimate the noise from the AOM actuator alone in gyro mode, taking into account the double pass. This agrees pretty well with (1) in the region we think is dominated by oscillator noise
5. This isn't really an oscillator phase noise measurement. It is the feeback signal to the PLL oscillator with the modulation on the AOM off (i.e., no gyro signal). This is significant in that, since it is roughly equal to the estimated PLL oscillator noise at low frequency, it rules out the possibility that the excess LF noise comes from phase noise in the output MZ. Therefore, the excess noise must be noise imparted on the CW light by the secondary loop. This helps narrow things down.

I measured eight of the 45P mirrors from the 40m.  The transmission doesn't look as high as we realistically need it to be.  They were: 57ppm, 38ppm, 46ppm, 36ppm, 48ppm, 64ppm, 34ppm, 55ppm, with some ~10ppm error.

Using combinations of these mirrors the best we could do is a transmission through the cavity of 20% with losses at 40% and a contrast defect of 0.22, the rest being lost in transmission through the turning mirrors.

We are starting to get quotes back in for coating runs.  It seems that they may not be able to get exactly the transmission we want, but instead will have +- 30ppm.  It may be that we want to be a bit cautious in our transmission choice since 100ppm -30 would end up with a transmission much lower than we want.  If we go for a higher transmission in the first place then we reduce that risk but may get a slightly lower finesse, for example 130ppm would give a finesse of 15k instead of 18k for 100ppm.  The extra upside is that we get a better contrast defect (0.12).  

The C_d is never going to be zero since the input and output mirrors will have very similar transmissivity.

I've put in requests for quotes to a number of companies now for 100ppm coatings.  I also checked through our history of measured transmissions, and notices that these mirrors have some that are very close to what we want.  The Y145P 2" mirrors that I previously measured seem to have much higher transmission for 45S than the ones in the Mott measurements.  I'm going to go across and dig these out again so we can repeat the measurement.

 [Alastair, Koji, Zach]

As we discussed doing, we removed the PBS that was used to split the power between beams and put a 50/50 power splitter in its place:

The thinking was that doing this would minimize the AM coupling from polarization rotation in the EOM. After realigning the entire experiment, we observed no improvement in the low-frequency noise spectrum (compare with this post):

It does seem like there is a reduction from 10 mHz to 100 mHz, but it is tough to tell whether or not this is meaningful given the FFT bandwidth. There remains the possibility of true AM from the EOM itself, though it seems unlikely that it would be stronger than that from the rotation. I suppose we should measure it either way.

Another not-extremely-likely case is that the phase noise from the two oscillators is the same below 1 Hz. This means that their absolute noise levels (V/rHz) would have to be different by a factor of two to accommodate the fact that the AOM oscillator noise couples in twice as strongly from the double-pass. Since the phase noise tends to go up with carrier frequency, and since the AOM carrier is half that of the PLL, this isn't out of the question. We are planning to beat our two Marconis together to see if the low-frequency noise is enough to explain the gyro noise. Frank and Tara's data suggest that it stays fairly flat at lower frequencies, but they don't have a measurement at the low frequencies we're concerned with.

 The polarization going into the EOM was off by 90 degrees. This must have happened when I was minimizing the DC offset at some point (since there is a minimum whenever we are aligned with either of the crystal's axes, but of course the phase modulation is minimal here). I fixed it and we now have reasonable sidebands:

The carrier peak is at ~11 V, the sidebands are at ~176 mV. Using the 1st order bessel approximation, the modulation depth is

Γ = 2 sqrt(0.176/11) = 0.25 rads        this is about what we want.

This is almost certainly what is happening---I guess I assumed that was the case. We can trim the offset that develops away by making slight changes to the HWP before the EOM, counteracting the rotation that happens in the modulator. I have seen lots of setups with PBSs after modulators, so I didn't foresee it being a problem when we designed it in the first place.

I guess we could use a power splitter instead of the PBS, then put a waveplate directly after it in the AOM path (before the second PBS that directs the beam to the AOM). Using the waveplate, we could make small changes to the power that gets sent into the AOM (vs straight through the 2nd PBS into a beam dump) to finely balance the power without having a maximal linear coupling from the polarization rotation in the EOM.

 To estimate the RAM before the measurement, we can use the following logic:

The cavity has a linewidth of ~100 kHz. According to the Gyro Doc the gyro signal df ~ omega / lambda, where omega is the rotation rate. Since the noise now is ~10^-5 rad/s, this is equivalent to ~10 Hz/rHz of frequency noise at 0.1 Hz.

This would require a AM/PM ratio of ~(10 Hz / 100 kHz) ~10^-4 at 0.1 Hz. Its easily possible; I have seen things of this size.

Another possibility is that our setup has been made especially sensitive to this by the PBS after the modulator. The RAM typically gets made by unwanted polarization rotation. If we have tuned a waveplate to split the power after the EOM 50/50, it means that we are also in the configuration where the polarization -> AM conversion is maximized. We can reduce this by trying to do a polarization insensitive split or by having separate EOMs in the two paths.

To verify that its a polarization to AM conversion, you ought to adjust the waveplate to put minimize the power in one path and measure the RAM in the other one. Then you ought to be quadratically sensitive to the angle.

Here is the current gyro noise spectrum, as computed from both the AOM and PLL control signals:

It is not any better than before, but there is one important thing going on here: the noise in the PLL signal is lower than that in the AOM signal between ~1-100 Hz. The noise in this band is dominated by oscillator phase noise (from the AOM driver or from the PLL LO, depending on the signal). This marks the first time that we can verifiably show that some noise in the locking loops is suppressed in the transmission readout. This has been predicted by our noise model for quite some time, but we haven't really been able to demonstrate it yet.

In fact, it's been there ever since we rebuilt the gyro, but I had forgotten a factor of two in the AOM calibration due to the double-pass. If you look at some older plots, you can see that the noise at high frequencies differed between the two signals by just enough to make the mid-frequency noise look the same (while even making the AOM noise look BETTER than the PLL noise at low frequencies).

I'm pretty confident that what we see now is correct. The noise above ~200 Hz is well described by the "spillover" noise, so the only mystery that remains is the excess low-frequency noise. It appears to be exactly the same in both loops, so this isn't something that is being suppressed by the CW loop gain. This has been verified by watching the LF noise as we turn the CW boosts on and off.

The main suspect at the moment is the RFAM drift we see as a modulation of the primary-loop error signal offset, so this will be our next focus. I will soon have a quantitative analysis of the AM/PM levels we are getting from the EOM, and drifts with time therein.

http://www.youtube.com/watch?v=IRzcoRFkpBQ

http://www.youtube.com/watch?v=hHN20qP0HCY

 The right way to choose the component values is to write down the correct cost function which you are trying to minimize. Since its a straightforward kind of circuit you can analytically construct the Jacobian.

The LMH6624 has a dynamic range which is larger than the what you can achieve with with the light (i.e. signal/shotnoise).

I didn't explain that properly in the last post.  We had set up the PD with a low gain because we need to keep the signal low enough for the mixer.  After putting it together and measuring the transfer function we noticed that there was a large peak at high frequency as Zach mentioned in this post.  Today I went back and remeasured it after removing the 200MHz notch, and then started increasing the gain to see if it would go away.  That is the reason why I have plotted the three different gains.

Definitely we need to keep the gain at >=10.  The issue now is whether we need to reduce the gain somehow (ie reduce the transimpedance using a smaller inductor, or finding a low gain opamp).  Any views?

There is no case in which you can run the MAX4107 or LMH with less gain than the minimum gain of the datasheet. If you look at the time series of the PD with low gain its probably oscillating like crazy.

From the datasheet's Bode plots, you can see how the phase margin is for low gains: bad news.

I thought about it, but only because of the rainbow font. And the plz.

so plz do not mess with the instruments attached to the setup

(don't even think about it)

After some major software issues (trial license had expired ) i re-installed everything.
Also characterized the first bunch of 3mm EG&G diodes from PK, so that i can have the not awesome ones for destruction.

Current device being tested  is #1136, 20W and 2ms pulse duration.
Taking measurements of impedance, dark current, dark noise etc. every 100 pulses.
Also taking pictures after every single pulse, 1000ms delay between pulses.

Yes.  Jokes on you fool!

Is this an April Fools joke?

Here are some new plots of the 100MHz transmission PD.  I removed the 200MHz notch that I had put in (since we don't need it), and saw that the resonance at high frequency looked pretty bad.  I wanted to check if this was because of the low gain we were using (datasheet says that it is stable for gain>=10), so I retook the TF and then changed the gain up to 3 and then 10 by changing R5 to 100 and then 453.

The gain3 is an improvement, but there's still some nastiness at high freqency.  The gain 10 plot looks pretty clean though.  Compared to the liso model (which I have set to gain 10, and have changed the diode capacitance down to 100pF) the gain seems to be a lot higher than we should be getting.  I've gone over the calibration a few times and don't see any errors.  I've also checked the values in liso and they seem correct too.  This is not great since we really don't want to have such a high gain in the first place.

We should probably see if there is a low gain version of the LMH6624 that we can use in its place.

EDIT:

I should also mention that I did take into account the difference in the laser power on the two PDs.  The 1611 had 1.16mW on it, and our PD had 0.47mW on it.

 I reinstalled the REFL PDs and relocked the gyro. I am going to have to wait for a brief pause in Frank's measurements tomorrow to tune the EOM circuit; it is running in broadband now. After everything is back in place I will take the new loop TFs, etc.

Attached again

i tried to open the liso file but it seems to be empty

 [Alastair, Zach]

Alastair finished building our first TRANS PD (S/N 03), and we tuned it and took a transfer function with the Jenne Laser:

Not sure quite what's happening up at 277 MHz. The gain at the readout frequency (100 MHz) is a bit higher than what is predicted by the model, but that's probably because Alastair used 200 pF as the diode capacitance, whereas it's closer to 100 pF with the 5V bias. 200 pF puts the diode pole (RC = ~10 ohm * 200 pF) at ~80 MHz, but it's probably more like 160 MHz, explaining the extra gain.

We also retuned the REFL PDs to have their readout and 2f rejection frequencies at 19 & 38 MHz, respectively. They match up quite nicely, save for some difference in the diode capacitance:

I still need to re-tune the EOM resonant circuit, which I will have to sneak in during a break in Frank's diode demolition derby tomorrow. Otherwise, we're ready to get the gyro gyroing again.

I'm in the middle of stuffing the first Trans_PD at the moment.  I'm using the following values for the resonant notches:

• AC coupling
  • L6=6.8uH
• 100MHz notch
  • C2=22pF
  • C4=9pF(1.5-15pF tunable)
  • L3=82nH
• 200MHz notch
  • C10=6.3pF(1.5-15pF tunable)
  • L2=100uH

You can compare these values to Zach's 33MHz diodes in this post.  The lower inductance of L3 gives a lower transimpedance and I've set the gain of the opamp to 2.

I have some concerns that we don't want to much gain for these PDs.  A total transimpedance of 30dB V/A means that 20mW of laser power becomes about 7dBm at the output of the PD.  If we're using a level 13 mixer then that seems about right.  If we want to increase the power at all then we will need to attenuate the output of the PD.

LISO file is attached

I updated arbcav to include more functionality. Now, in addition to giving you the transmission/reflection spectrum of the cavity you tell it to build, it will also calculate the HOM structure (carrier and sideband) for linear, triangular, and quadrangular cavities, taking astigmatism from arbitrary angles of incidence into account. You have the option of whether or not to include RoC for the mirrors. If you don't, it will just spit out the T/R spectrum as before; if you do, it will give the HOM info. You can also choose whether or not to give it a modulation frequency, and if you don't it will only plot the carrier HOMs.

I've updated the copy on the SVN (link above).

Below is the output of the function for our cavity. With the new length of 3 m (for FSR = 100 MHz), it looks like f_mod = 19 MHz is the best bet.

I have made the HOM plot such that higher orders correspond to lighter shades (as well as shorter lines) to aid in identifying the stronger modes.

I guess that this is a transfer function between the source out of an RF analyzer and the RF output of the New Focus 1811 PD (which goes up to 125 MHz). If so, It looks like the EOM is nicely free of mechanical resonances in the band we care about, Nice job, Thorlabs.

Here is the RFAM measurement redone.  First plot is the full spectrum from DC to 100MHz.  Second plot is put together from multiple scans to give higher resolution in the area of interest.  We used the same setup as before, but with an 1811  instead of the PDA255 to improve bandwidth.  The steps in the plot at 15MHz and 25MHz are the points where the scans have been joined.

Thanks for adding that Joe.

I've changed the description in the ATF wiki to include restarting the front-end in the MEDM screen, and I've added a description about how to get the channel status from daqd using telnet, since that's a handy thing to be able to double check.

At Rana's suggestion, the GYRO_MAIN medm screen background is now linked to the C2:DAQ-FB0_ATF_STATUS epics channel, so that the whole screen turns a horrible pink/red colour if this channel goes to anything other zero.

[Joe, Alastair, Zach]

Since I haven't seen an elog about it, I'm putting it in now.

What happened:

Yesterday, I visited at the request of Alastair.  During the LVC meeting it was noted there was a DAQ configuration mismatch between the frame builder process (daqd) and the front end (c2atf).  There was an error message, 0x2000, on the C2ATF_GDS_TP.adl screen indicating this.

In this case I simply hit the reload DAQ button on that screen and it made the front end and frame builders happy.  The channel label in dataviewer went from being RED to being the usual black.

In addition, Alastair also pointed out the daqd often times crashed when starting up because it was restarting itself too quickly after crashing.  To remedy this, I went to the /cvs/cds/caltech/target/fb directory and modified the daqd.inittab file and added a line "sleep 5" near the beginning. 

We did a restart of the frame builder by using "telnet fb0 8087" and then typing "shutdown", and found it came up perfectly fine.

Lessons to be learned:

First, realize that for data to be recorded by the frame builder,  both the frame builder (the daqd process) and the front end (the c2atffe.ko, which can be seen to be running by typing "lsmod" on the front end computer) need to have loaded the same version of the daq configuration file. 

This file lives in /cvs/cds/caltech/chans/daq/ and is called C2ATF.ini.  So whenever you change this file, whether thats by rebuilding the front end, running daqconfig, modifying it by hand, or running a script to modify it for you, you need to:

1)Restart daqd ("telnet fb0 8087", then "shutdown")

2) Force the front end to reload the file by either pressing reload DAQ or by restarting the front end code (using /cvs/cds/caltech/scripts/startc2atf)

 [Alastair, Zach]

We finished re-setting up the gyro this evening. Alastair cut a hole in the top of the side of the IO box so that the lid can be put on with the necessary cables fed in. Everything, including the PLL, is back online with everything working from the rack.

I had some of the time-varying error signal offset issue coming back, but I was able to get rid of it by iteratively adjusting the HWPs before and after the EOM so that 1) there was minimal RFAM and 2) the power was balanced between the CCW/CW beams. I also adjusted the modulation frequency again slightly because it looked as if we were sitting on a bad spot with respect to the modulator's AM/PM response. We still need to take a proper transfer function of this (with better resolution than the old one) to be sure, and once we're done settling on a frequency we have to retune the REFL PDs and EOM resonant circuit.

I will continue with the loop characterization tomorrow.

 [Alastair, Zach]

As planned, we have reorganized the instrument placement and signal routing to be cleaner and more efficient. The general structure is to have nearly all of the instruments (FGs, servos, supplies, filters, etc) in the rack under the monitors, with one BNC patch bay dedicated to carrying signals to and from the experiment. A separate bay provides access to the DAQ from the rack for acquisition/excitation. A single cable runs from the DAQ to the experiment to provide the slow control for the laser. We opted to have the AOM amplifier on the table to minimize the distance the ~6W of electrical power has to travel to the modulator.

Four channels between the table and the rack are dedicated "scope" channels; this way, we can have any of rack-bound signals viewable on the scope at the table while futzing with the experiment. 

As you can see, this already leaves a lot of space on the enclosure shelf for anything we may HAVE to put there in the future. The power supplies visible in the picture will also be moved elsewhere.

Also, we checked an unused Marconi out of the 40m so that we have one for each the AOM and the PLL. I think we were supposed to do this a while ago.

Visitors to the lab ...

 Did you change anything since then in the front-end code?  I'm looking at it right now, and it looks like a physical wiring problem not the front-end code.

 Due to an error somewhere in the front end code, the channel that should have been "PLL_ACT" was actually receiving the "CCW_REFL_DC" signal. All data and plots for noise in the PLL actuation readout are absolutely wrong.

Having made sure that I was monitoring the right damn channel, I observe that the PLL actuation noise is once again the same as the AOM actuation signal. This puts us at ~2 x 10^-5 (rad/s)/rHz at our most lenient requirement frequency of 200 mHz, or roughly 2700x where we need to be. I currently have and all enclosures shut and all boost engaged (2 for the primary loop and 1 for the secondary). Changing the number of boosts does not affect the low frequency noise relationship between AOM_ACT and PLL_ACT. The secondary loop has >60 dB of OL gain below 50 Hz, so if changing the boosts does not affect the noise then it means that either 1) the same noise that couples into the AOM signal is coupling in independently into the PLL loop and causing the same noise level despite the suppression from the secondary loop, or b) we don't understand how it really works. I cannot see how differential IO noise is not suppressed by the secondary loop gain as measured in the PLL actuation; this just doesn't make sense.

I'm not really sure what else to do. We have isolated all the optics from the open air, and ensured that the locking loops have a pretty solid amount of gain (the secondary loop gain is not astronomical, but it's enough that we should see something) and yet we can't reduce the noise at the transmitted port. The addition of the vacuum system and custom PDs has amounted to a factor of ~5 better noise. Big whoop.

Also, the chamber pressure is now up to about 1.7 torr (no factors of 10 here).

 I've finally got round to taking the photos off my camera, so here are some of the new parts made for the gyro. 

Stainless steel mounts replace the old brass ones for the in-vacuum optics.  They needed to be shorter anyway, since they are sitting on the flange inside the vacuum, so we took the opportunity to try to increase their stiffness.  I'd have liked to get another dogclamp on them in the vacuum, but as you can see the space was limited.  All the in-vacuum mirrors are now mounted in SL mounts.

I've added a picture of the new diodes on their bases.  These have brass bases, again with the dogclamp and 3 foot design. There is a black plastic spacer in there too to give electrical isolation from the bench.  At this point we hadn't fitted the aluminum heatsink/clamp for the PD.

Ah, "- This AOM feedback should agree with the calibrated error signal of the secondary cavity. This is not confirmed." was

The AOM feedback (calibrated) with the loop closed = The secondary error (calibrated) without the loop closed

"- This noise is suppressed by the loop gain of the secondary cavity. (The spillover noise of the secondary.) This is not shown. Also the OLTF of the secondary is not shown."

The spillover noise exist for the primary and the secondary.

The measurement with the beat signal relys on the fact that each laser frequency is following the resonance of the cavity.

In your case the real gyro signal is way smaller than the PDH error felt by the cavity because of the differential input optics or whatever.
If this noise is not suppressed by the servo, the noise transmits the cavity and felt by the beat. Therefore,

We need to know
1) The free run secondary error level
2) The suppressed secondary error
3) The OLTF of the secondary

"Here are both of the OLTFs:"

The difference between the AOM feedback and the beat measurement is ~50. This is 34dB, while the OLTF below 100Hz is larger than 50dB.
So presumably, the spillover noise is smaller than the measurement. This is a minimum confirmation of the secondary spillover noise.

I'm not sure I understand your points..

"- This AOM feedback should agree with the calibrated error signal of the secondary cavity.  This is not confirmed."

I think you mean that the secondary error signal calibrated into Hz via the optical response ([V/Hz]) should be the same as the suppressed noise calculated by multiplying the AOM feedback signal through the AOM response ([Hz/V]) and dividing by the OLTF, as I did for the primary loop here. If so, then no, I haven't done this yet. I was about to do it now but I must have left the CCW error signal disconnected from the DAQ 

"- This noise is suppressed by the loop gain of the secondary cavity. (The spillover noise of the secondary.) This is not shown. Also the OLTF of the secondary is not shown."

I don't think I know what you mean here. The spillover noise from the primary cavity (which can be calculated from the primary error signal) is present in the AOM actuation signal, as it tries to cancel the noise that the primary loop has failed to.

Here are both of the OLTFs:

"- The AOM feedback at around 100Hz shows the same level as the primary spillover. Does this suggest that the loop gain of the secondary is higher than 1k???"

In frequency bands where the residual common-mode noise is higher than the differential noise, the AOM feedback spectrum should be the same as the calculated spillover noise (blue and red, respectively). This appears to be true above ~100 Hz. It was true above 30 Hz in the old budget. It makes sense that the residual common-mode noise has gone down a little below 100 Hz, because the primary OL gain has gone up a bit from the extra integrator I added, but it is strange that there is still about the same level of noise below 100 Hz in the AOM signal as there was before, despite the fact that we replaced PDs. It could be that the optical gain is still low enough that electronics noise dominates here, but we won't know that until I plot the PD/demod noise alongside it. I will do that tonight or tomorrow.

"- Of course we like to see the OLTF of the PLL."

I plan to measure this, but it is a little difficult since I don't have an electrical summation point for this as I do for the locking loops. I think I can do it by sweeping the AOM frequency (with the secondary loop open, that is), but I haven't worked it out yet.

I understand your explanations. Now I feel these points.

- This AOM feedback should agree with the calibrated error signal of the secondary cavity.  This is not confirmed.

- This noise is suppressed by the loop gain of the secondary cavity. (The spillover noise of the secondary.) This is not shown. Also the OLTF of the secondary is not shown.

- The AOM feedback at around 100Hz shows the same level as the primary spillover. Does this suggest that the loop gain of the secondary is higher than 1k???

- Of course we like to see the OLTF of the PLL.

Are you working with the measurements, Zach?